Our senses aren?t passive. Rather, we actively seek relevant information via sampling movements. However, experiments in sensory systems often restrict sampling movements to simplify stimulus delivery and allow large scale imaging and electrophysiology. But whether sensory systems function equivalently in restraint vs free sampling remains an open question. The mouse olfactory system presents a perfect opportunity to address these issues. Mice are in constant motion, actively sniffing throughout their environment. Here, we will establish an array of innovative techniques to image the sensory input to the olfactory system, and to control that input optogenetically, all in unrestrained mice. We will perform these experiments in the context of an unrestrained olfactory navigation assay we have established. Mice learn this task rapidly, and solve the task by following the odor concentration gradient. Using real-time video tracking of nose movements and thermal recording of respiration, we can monitor their sampling strategies. We have found that mice develop a consistent repertoire of movements that are precisely synchronized to the sniff cycle. Importantly, these behavioral dynamics are only revealed when an animal is able to move naturally in the environment. In aim 1, we will image sensory input to the glomeruli of the olfactory bulb during this navigation task. In mitral and tufted cells, we will express genetically encoded glutamate indicators, which have much faster kinetics than the more widely-used calcium indicators. We can thus capture fast temporal patterning of the input to the olfactory bulb. To image in freely-moving mice, we will implement a miniature microscope design that can image the entire dorsal surface of the olfactory bulb in freely-moving mice. In aim 2, we will develop a method for driving navigation behavior with fictive plumes that we control optogenetically. Using a mouse line that expresses ChannelRhodopsin-2 in olfactory sensory neurons, we will deliver light stimuli in closed loop with the mouse?s movements and inhalations. In the timeline of this grant, we will use this technique to create fictive gradients of input amplitude, a neural coding cue that is thought to represent odor concentration at the glomerular level. We have established a behavioral task that allows us to study active sampling during olfactory navigation. In this proposal we will perform the initial studies using this framework to understand the underlying neural mechanisms. This work will advance our understanding of how we optimize our sensory input through sampling behavior, a process that goes awry in neurological disorders such as schizophrenia and autism.